Research data
and downloads.
Schema 1.0. These static, cacheable files are generated from the same public historical-person model as the human catalogue. Start with the compact resolver index, then fetch one dossier or depth-one family network.
Current data and small downloads.
Build manifest (JSON) · Build 18a178c3802cd896e24c · Assembled 2026-10-07T15:21:37+00:00. Latest dated person research or review: 2026-10-07.
Start with the small indexes below, then follow each person’s JSON link for the current resolved dossier. Approved reviews override imported fields; imported preserves the original values and provenance. Unreviewed fields retain their existing qualifications. The bulk people.json export is an imported research snapshot, not the reviewed interface.
The manifest records dataset scopes, entry counts, byte sizes and SHA-256 hashes. Re-fetch it when resuming research; a changed build identifier means the listed data changed. Build time is separate from research dates.
Browse small JSON chunks
- People: 4,800 entries — 1 · 2 · 3 · 4 · 5 · 6 · 7 · 8 · 9 · 10 · 11 · 12 · 13 · 14 · 15 · 16 · 17 · 18 · 19 · 20 · 21 · 22 · 23 · 24 · 25 · 26 · 27 · 28 · 29 · 30 · 31 · 32 · 33 · 34 · 35 · 36 · 37 · 38 · 39 · 40 · 41 · 42 · 43 · 44 · 45 · 46 · 47 · 48
- Records: 4,877 entries — 1 · 2 · 3 · 4 · 5 · 6 · 7 · 8 · 9 · 10 · 11 · 12 · 13 · 14 · 15 · 16 · 17 · 18 · 19 · 20 · 21 · 22 · 23 · 24 · 25 · 26 · 27 · 28 · 29 · 30 · 31 · 32 · 33 · 34 · 35 · 36 · 37 · 38 · 39 · 40 · 41 · 42 · 43 · 44 · 45 · 46 · 47 · 48 · 49
- Families: 353 entries — 1 · 2 · 3 · 4
Choose a starting point.
Predictable endpoints
| Purpose | Path |
|---|---|
| Compact global index | /data/people-index.json |
| Person dossier | /people/<lowercase-wikitree-id>.json |
| Depth-one network | /people/<lowercase-wikitree-id>.network.json |
| Exact normalised-name candidates | /data/resolve/<name-slug>.json |
The static resolver returns candidates, never a genealogical conclusion. Filter or score candidates using birth year, all locations, spouse names, parents, children and occupations in people-index.json. The generated score is deterministic: normalised name, birth-year tolerance, location and spouse matches add explainable reasons; conflicts subtract a documented amount.
Claims, evidence and relationships
The dossier follows Person → Claim → Evidence → Source. Stable hash-based IDs prevent rebuilds from renumbering unchanged objects. Canonical claim statuses are unknown, possible, probable, strongly_supported, proved, disputed and contradicted; the original project label is retained in status_label. Source quality is original, derivative, secondary, tree_only or unknown.
A relationship with tree_relationship: true says the exported WikiTree currently connects the people. It is not documentary proof. Children with relationship_source: reconstructed_parent_reference were recovered by reversing child profiles’ Father/Mother fields. A proved relationship carries claim and evidence IDs from the controlled evidence assessment.
Research questions and leads
open_questions come from versioned local research cases. research_leads include pending case targets plus conservatively extracted profile-note actions only when they contain a concrete archive reference, URL, or a record type with a constrained year. Repository, collection/reference, date bounds and reason are retained where supplied. No archive references are invented.
Provenance and interpretation
- Mapped associations, WikiTree tree links, extracted citations, controlled relationship assessments and local research findings retain separate provenance labels.
independence_groupgroups citations that resolve to the same underlying URL or source key.- Estimated dates and locations remain explicitly labelled.
- A cluster is a research grouping, not proof of kinship.
- Full profile captures remain available in
wikitree-profile-evidence.json, but compact dossiers omit raw biographies.